How to start a loyalty program for a small business
Most loyalty programs go wrong at the first decision, not the tenth. Six choices, in order, from what to reward to the first message you send.
On this page
- 1. Decide what you are rewarding
- 2. Pick the reward and the card length
- 3. Choose the mechanism: paper, app, or wallet pass
- 4. Design the card
- 5. Get people onto it before you do anything else
- 6. Send the first message, and only one
- Common mistakes when starting
- Frequently asked questions
- Where to start this week
Most owners who ask about starting a loyalty program are really asking about the card: what it should look like, what the reward should be, which app to use. Those come later, and they matter less than the question usually gets first.
The question that actually decides whether the program works is simpler and gets skipped: what are you rewarding, and how will someone join without thinking about it? Get those two right and the rest is mechanical. Get them wrong and a well-designed card still sits unused in nobody’s wallet.
This is the order that holds up in practice, one decision per section, each one feeding the next.
Key takeaways
- Decide what you are rewarding before you decide the reward. Visits and spend pull customers toward different behavior, and most independent shops want the first one.
- Pick the card length from how often customers actually visit, not from a round number. How many stamps should a card have? has the arithmetic.
- A wallet pass beats paper and an app on the one thing that decides whether a program survives: it reaches a phone that is already in someone’s pocket, with nothing to download and nothing to lose.
- Enrollment is the gate. A beautifully designed card with 40 members cannot move anything; a plain one with 400 can.
- Send one message before you send a second. A program’s first week should prove it works, not prove how much it can do.
1. Decide what you are rewarding
Before the card, before the app, before the reward: are you rewarding visits or spend?
A visit-based card gives a stamp per transaction, regardless of size. A spend-based card gives a point per pound. They train different behavior.
A stamp card rewards a quiet-Tuesday £3 coffee exactly as much as a busy-Saturday £3 coffee. That is usually what an independent shop wants: more visits, spread out. A points-on-spend scheme rewards the big ticket instead, which fits a boutique or a higher-average-order retailer better than a café or a barber.
Most independent shops, cafés, bakeries, salons, barbers, pubs, takeaways, run on frequency rather than basket size, so a visit-based card is the usual default. Stamps, points or vouchers covers the full decision, including when spend-based is actually the better fit.
2. Pick the reward and the card length
Once you know what you’re rewarding, the card length follows from how often your customers already visit, not from a round number like 10. A shop whose regulars come weekly wants a card that completes in six to eight weeks, not one that stretches to three months before anyone sees a reward.
The reward itself should cost you something you can afford at the volume you’re hoping for, and should be the actual product, not a discount: a free coffee, a free cut, a free loaf. How many stamps should a card have? walks through sizing both the card and the reward against your own visit frequency and margin.
3. Choose the mechanism: paper, app, or wallet pass
Three ways to run the card itself, and they are not close to equivalent.
Paper costs nothing to start and nothing to keep running, and that is also its ceiling. A paper card cannot message anyone. It gets left in a drawer, lost in a wash, and tells you nothing about who your customers are once they walk out the door.
A downloaded app solves the messaging problem but adds a step most people will not take. An app store listing, a password, and a reason to open it again are three places for someone to quit before they ever earn a stamp.
A wallet pass (Apple Wallet or Google Wallet) sits in software already on the customer’s phone. There is no store listing and no account to create, and the pass can still carry a message to a lock screen the way an app can.
Free loyalty cards breaks down what paper and the free app tiers actually cost you in practice. The complete guide covers how a wallet card works end to end.
| Mechanism | What it costs | What it gives up |
|---|---|---|
| Paper | Nothing to start | No messaging, no data, easy to lose |
| A downloaded app | A developer or a subscription | An app-store install most people won’t make |
| A wallet pass | A paid plan, usually | Little: it’s already on the phone, nothing to download |
A wallet-pass loyalty program is also the base a business can build on later. The same pass can carry push notifications, and email or text can follow up once a message needs more room than a lock screen gives it.
Passumo runs all three from one card. That matters less at launch than it does once the program is live and you want to reach people who have gone quiet.
4. Design the card
A wallet pass has a small, fixed layout: a few fields, a logo, a progress indicator. There is not much to get wrong, but the few decisions that exist carry almost all of the result.
How to design a loyalty card people keep on their phone covers the three that matter. Loyalty card examples is worth a look before you start: four real cards taken apart for what makes people actually use them.
Resist the urge to spend a long time here. A clear, simple card that ships this week beats a polished one that is still in review next month.
5. Get people onto it before you do anything else
This is the step that decides whether any of the above matters.
Enrollment gates every other number in a loyalty program. A card with 40 members cannot move a quiet Tuesday no matter how well it is designed. A plain card with 400 members can, because the program’s whole value comes from being able to reach people, and you cannot reach someone who never joined.
For a physical shop, the counter at the till is the entire signup flow. There is no landing page and no ad, just a code and a few seconds of attention while someone is already holding their phone.
Scanning that code is not a fading habit either. Bitly’s 2026 scan report recorded QR scans growing faster than new codes being created, across every region it measured, including North America (Bitly, published March 2026). What to put next to the till covers where the code goes, what the sign should say, and why the one sentence staff say at the till tends to outperform the sign itself.
Ask for less than feels natural on the join form. A name is usually enough to start, and a name-only member is still reachable through their wallet pass. Marketing consent for email and text is a separate matter, and it needs to be an explicit choice, not a box that is already ticked. Getting opt-in covers what UK law requires, and why asking properly also gets you a list that actually responds.
6. Send the first message, and only one
Once people are joining, the instinct is to build out a full campaign calendar immediately: a welcome message, a win-back sequence, a birthday offer, a monthly newsletter. Leave all of that for week two.
Send one message first: a plain confirmation that the card is live and what the reward is. Watch what happens. Does the push land, does anyone redeem, does anyone complain.
A program’s first week should prove the basic mechanism works before it proves how much it can do. Push, email or SMS: which one to send covers which channel that first message belongs on. Once the basics are confirmed, automated campaigns covers building out the welcome sequence and the rest without doing it by hand every time.
Common mistakes when starting
Choosing the reward before the metric. Deciding “free coffee after ten” before deciding whether visits or spend is the thing worth rewarding tends to produce a card that fits the business’s assumptions rather than its regulars’ actual behavior. Decide the metric first.
Spending the first week on design instead of enrollment. A card that looks right but that nobody has joined does nothing. Ship a simple card and put the real effort into the sign at the till.
Collecting too much on the join form. Asking for a birthday, a phone number, and an email before handing over a card loses people who would have joined for a name alone. Add fields later once someone has a few stamps and a reason to trust you.
Running every campaign type in the first month. A new program with no track record of what its own customers respond to is the wrong moment to automate five message types at once. Prove one works, then add the next.
Frequently asked questions
How long does it take to actually get a program running? The decisions above take an afternoon. Designing a wallet card and printing a counter sign is usually the same day. The slow part is never the setup; it is building enough members for the program to be worth messaging, which takes a few weeks of consistent asking at the till.
Is it worth doing for a single shop, or only once I have more than one? Worth doing for one. The arithmetic that makes a loyalty program pay, more visits from people you can already reach, works the same at one site as at five. Whether to run one card across two sites once you open a second is a separate question, worth revisiting once you get there, not before.
Do I need an app to run this? No. A wallet pass needs no app download from the customer; they already have Apple Wallet or Google Wallet on the phone. The complete guide covers how the wallet mechanics work.
What does it cost to run? Paper is free and gives you nothing to message with. A wallet-pass platform is usually a paid plan; see the pricing page for current numbers. Free loyalty cards covers what the free options give up to stay free, so you can weigh the trade honestly against a paid one.
What should I measure in the first month? Not scans or downloads. Measure how many members you have relative to your transaction count, and whether the first message you sent got opened and redeemed. Everything after that, win-backs, birthdays, segmentation, is easier once that basic loop is confirmed working.
Where to start this week
In order: decide what you’re rewarding, size the card to how often people actually visit, build it as a wallet pass, and put the sign at the till before you touch anything else. The messaging can wait until people have actually joined.
Passumo builds the wallet card, the join page, and the first message in one place, with push, email and SMS available from day one whenever you’re ready to use them. There are worked examples for cafés, for salons, and for restaurants.
Source: Bitly, The State of QR Code Scans in 2026, published 10 March 2026, drawing on roughly 2 billion scans on the Bitly platform. A vendor publication rather than independent research, and the sample skews toward codes created on a paid platform.